Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×
Medicine

FDA Warns Against Using Smartwatches and Smart Rings To Measure Blood Sugar (cnn.com) 50

In a warning issued Wednesday, the FDA said it has not authorized or approved any smartwatch or smart ring to measure blood glucose levels. The use of these devices can lead to inaccurate measurements and errors in managing diabetes that can be life-threatening, the agency said. From a report: These unauthorized devices are different from smartwatch apps that display data from FDA-approved continuous glucose monitoring devices that pierce the skin. The FDA did not name specific brands but said the sellers of these unauthorized smartwatches and smart rings advertise using âoenon-invasive techniquesâ to measure blood glucose without requiring people to prick their fingers or pierce their skin. However, these devices do not directly test blood glucose levels, the agency said, urging consumers to avoid buying them for that purpose.

The agency also advised health care providers to discuss the risk of using unauthorized blood glucose measuring devices with their patients and to help them select an appropriate authorized device for their needs. âoeThe agency is working to ensure that manufacturers, distributors, and sellers do not illegally market unauthorized smartwatches or smart rings that claim to measure blood glucose levels,â the FDA said in the statement. âoeIf your medical care depends on accurate blood glucose measurements, talk to your health care provider about an appropriate FDA-authorized device for your needs." .

This discussion has been archived. No new comments can be posted.

FDA Warns Against Using Smartwatches and Smart Rings To Measure Blood Sugar

Comments Filter:
  • Well, you'd think as a type 1 diabetic and someone that values evidence-based medicine, I'd be all over the FDA warning people to know what they're doing. Unfortunately, though, I remember what the FDA had to say about my first continuous glucose monitor: 'don't use this product to make treatment decisions. Only use blood glucose test strips for decisionmaking.'

    So why the fuck was it on the market in the first place? As the world's worst and most expensive temporary tattoo? So I can look at random 2- and 3
    • Re:FDA credibility (Score:4, Informative)

      by thegarbz ( 1787294 ) on Thursday February 22, 2024 @08:49AM (#64259464)

      So why the fuck was it on the market in the first place?

      And why wouldn't it have been? It's not a treatment and the FDA hadn't approved it as such. That doesn't mean people can't use something.

      And more to the point, what the fuck happened to make them change that advice?

      People used the device to build up a wealth of data showing it was reliable. Now if it had not been on the market then this entire process would have taken significantly longer. Is that a positive health outcome for you?

      but now the treatment group could use that information, but the control group still couldn't? What the actual fuck?

      I hate to ask the obvious: Do you understand what a control group is? The "control" part of the name is as important as the "treatment" part of the other group. If either starts acting like the other it can invalidate all data gathered.

      So to answer your entire post: FDA has good credibility, they understand the importance of a strict scientific process. You on the other hand seem to lack this understanding.

      • by HiThere ( 15173 )

        They've made a few decisions recently that damaged their credibility with me....but in the direction of approving things on an other than experimental basis that didn't have sufficient evidence to warrant that approval.

        OTOH, they've also removed some generic drugs from generic status, and made them only available by trade name, at a much higher price. I can find NO excuse for that. (This didn't change their prescription status, only the price and who could manufacture them.)

  • Instead of scaring people, how about promoting innovation. I'd rather the FDA fund CGMs that work using light to sense (whatever) through veins than a permanetly mounted sensor up my arm.

    Yeah, I'm a diabetic. I use a CGM. FDA, do something usefu.

    Fuck these republican asshats.

    • by Entrope ( 68843 )

      The problem is that blood glucose fraction is so small that transdermal measurement is extremely hard -- and there's a lot of person-to-person variation in the raw data that device makers would like to use. The FDA probably would like it if somebody could show that their device works well for this, but nobody has done so.

      A big part of the problem are the rules for "works well". Devices with this basic function [fda.gov] (in that case, penetrative rather than non-invasive) are Class III, requiring premarket approval

      • by jonadab ( 583620 )
        > And the effective standard is basically that the device has to
        > be better than the existing standard of care in every respect

        If that were actually true, there's absolutely no way the Freestyle Libre (the kind where you stick a sensor to your arm that's supposed to work for a couple of weeks) would ever have been approved, because it's *way* less accurate, and in dangerous ways, than the more traditional strips. In particular, if it can't get an accurate reading, it doesn't show an error, it just ma
    • I'm not sure what the grant landscape looks like for optical CGMs, or related technology, specifically; but I don't think that the FDA does much medical research funding. Lots of food safety and some laboratory standards and development for the sort of wide-scale testing that food safety requires; but for drugs and medical devices they are mostly just in charge of judging clinical trials and postmarket reporting; not running their own.

      The Feds as a whole do a fairly substantial amount; but FDA research m
    • Instead of scaring people, how about promoting innovation.

      The FDA isn't scaring anyone. They are warning people of devices which have not been certified for use. This is a good thing if that device has health outcomes for you.

      As for promoting innovation, that isn't their job. Innovation has nothing to do with the FDA, and their lack of involvement doesn't change the fact that R&D into non-invasive measurement is alive and healthy, and once enough science is done showing that the equipment can be relied on they may even get that lovely FDA stamp of approval.

    • by necro81 ( 917438 )

      I'd rather the FDA fund CGMs that work using light to sense (whatever) through veins than a permanetly mounted sensor up my arm.

      The FDA is does not fund R&D. They're a regulator. Their federal appropriation doesn't have a line item for "fund research to make medical breakthrough X." A lot of their funding actually comes from "user fees" - the companies petitioning to bring drugs and devices to market. Sure, the FDA has some ways to "promote innovation" by easing certain rules and speed approval of

      • by jonadab ( 583620 )
        DARPA is mostly interested in developing technologies with the theoretical potential for military applications, although admittedly they are known to interpret that somewhat loosely on occasion.

        The NIH on the other hand really does seem like a good candidate for providing the kind of research funding the other poster was proposing. Seems like it would be right up their proverbial alley.
  • Tech companiers touting some cheap solution to a complicated medical diagnostic procedure that doesn't quite work, and might be dangerous if people take it as read? Surely that never happens?
  • It seems like reasonable caution to warn people who's lives depend on a reliably accurate measurement. The devices should have a warning as well and I'd be surprised if they didn't already to avoid liability. If I was a diabetic I'd probably use it and continue previous testing methods to compare accuracy before relying on it. A real curveball would be if it was more accurate, I'm not sure how one would not misjudge that comparison.
    • I assume that (barring some of the really fly-by-night ones who focus on being gone by the time anyone goes to investigate; rather than just being technically legal) do sport some variation on the "Not Intended to Diagnose, Treat, Cure or Prevent Any Disease" quack-Miranda you see on every "dietary supplement".

      It seems likely that the FDA is concerned(probably not entirely unreasonably) that the 'it's not a medical device; it's just for quantified wellness!" tech guys, mostly harmless when selling new nu
    • "The devices should have a warning as well and I'd be surprised if they didn't already to avoid liability."

      I'm sure they do. I'm also sure that the warning is in teeny-tiny font buried under a dozen pages of boilerplate legalese warnings.

    • by HiThere ( 15173 )

      The thing is, even the standard glucose monitors seem (to me) to have a lot of variation. I attribute this to low manufacturing standards for the (disposable) sensing device. (OTOH, I'm not about to rely on a smartwatch instead.)

    • Measuring just two methods you cannot determine which is more accurate. You could do two blood drop measurements and compare them.
  • Find 500 people with diabetes and give them a device for free. Then they keep a diary: Take a drop of blood and test it, take what the smart watch says. Check for differences.
    • by Entrope ( 68843 )

      No [fda.gov], it's not remotely that easy.

      • Yes it is and pointing to the FDA rules on a topic of how the FDA sucks is exactly making that point. There should be no reason that a representative series of studies on the subject canâ(TM)t be considered sufficient. The cost of FDA rubber stamping is expensive: https://www.fda.gov/industry/f... [fda.gov] and these annual fees are what has to be considered in the cost of making these devices.

        https://www.sciencedirect.com/... [sciencedirect.com] -
        The mean absolute relative difference of currently available glucometer models ranges

        • by Entrope ( 68843 )

          There should be no reason that a representative series of studies on the subject canÃ(TM)t be considered sufficient.

          The word "representative" there is doing an awful lot of work that the page I linked to only starts to outline. The FDA presumably has a much higher standard for what it thinks is "representative" than you do.

          • The FDA presumably has a much higher standard for what it thinks is "representative" than you do.

            Yes, only big bags of money given to "representatives" move the FDA.

          • by guruevi ( 827432 )

            Again, the highest scientific standard on a subject is peer reviewed published studies together with peer reviewed and published meta-analysis. That standard has been met and says that chemochrome sensors in wearables are as good and actually all show to be on the lower end of the error rate than 'classic' strips.

        • Yes it is and pointing to the FDA rules on a topic of how the FDA sucks is exactly making that point.

          The rules exist to ensure that the study is representative. You can always go back to the pre-FDA world where drug manufacturers actively caused you harm, sold you snake oil, or sold you medical devices that can't do shit.

          Why are you so keen to spend money on something that may not help you based on the word of a person who stands to make money alone, but against a slight increase in certainty all because the process raises the cost of a device? This isn't a fucking cheeseburger, it's something that has pot

          • by guruevi ( 827432 )

            Again tard, you think all people are as stupid as yourself? Before the FDA drug manufacturers caused less harm in comparison with the scientific evidence available at the point than today through the FDA which shields them from liability.

            I already pointed at a study, there are published meta-analysis on this particular subject that declare it as safe or better than current FDA approved solutions. Why do you think the government needs to put its finger on the scale on this subject?

            • You seem to know nothing on medical device testing. Others have pointed this out to you and yet you refuse to acknowledge your ignorance. This is a sign of stupidity, and adding the fact that you are seeming to get overly upset just confirms this. You are a moron in this area.
              • by guruevi ( 827432 )

                Actually, I do know a bit about FDA testing and medical devices, having designed programs for and work with them regularly.

    • That's the general concept of device testing, but it's more complicated in practice. The FDA has been doing this kind of thing for a long time, and they've seen all kinds of mistakes in the testing process that they'll want to avoid. To do that, they'll ask some more questions, such as:

      • How did you decide on the number of people in your study?
      • How long will the study last, and why did you decide on that study length?
      • How do you make sure the people in your study are representative of diabetes patients as
  • These smartwatches and smart rings have probably been pre-tested to a certain amount of accuracy, otherwise consumers will sue their pants off.
    • by necro81 ( 917438 ) on Thursday February 22, 2024 @09:53AM (#64259630) Journal

      These smartwatches and smart rings have probably been pre-tested to a certain amount of accuracy

      Well, there's testing, and then there's testing. (I have worked in the medical device industry.) You can do a lot of pretty informal testing and think that you have a good product. The FDA will look at that and say "Ok, this stack of test results and $2 will buy you a cup of coffee." In other words: all that informal testing doesn't count for shit, because it's easy for that kind of testing to mislead or be manipulated.

      Unless you have a well-designed clinical trial, and have put in place many measures to ensure 1) you are testing in real-world conditions, 2) you are testing a representative sample of users, based on 3) well-defined selection criteria and indications for use, 4) have something to measure against, like an existing device or industry standard, 5) have pre-registered your test methods and analysis plan, so you can't change the rules and metrics mid-stream, 6) ideally are double-blinded, and 7) report all your data, not just the most favorable, the FDA will not give you the time of day, and tell you to take a hike.

  • If they can't sell them, why should anyone else?
  • I think that just about summarizes the current incarnation of the FDA.
    • by necro81 ( 917438 )

      I think that just about summarizes the current incarnation of the FDA.

      For the cynics that wonder "why does nearly half of the FDA's budget come from 'user fees' paid by very companies that it regulates/", this article [uconn.edu] provides some background. The history is not as clear cut - with easily identifiable villains - as one might expect.

  • by gurps_npc ( 621217 ) on Thursday February 22, 2024 @11:13AM (#64259888) Homepage

    As a diabetic I assure you that it would be incredibly stupid to rely on devices like this. Even the approved devices need to be calibrated often.

    I can also assure you that there will come a time when your approved device is not available. They have parts that need to be replaced every 10 days, sometimes they break, and it can take days to get replacement parts.

    Not to mention issues when insurance changes.

    Finger pricks are still used to calibrate them and are great for meals when the devices are not working. But having a back up device to help you check your blood sugar will let you exercise without fear of accidentally committing suicide.

    • Bingo.
      I was thinking the same thing. If it is all you have, it is PROBABLY better than nothing, especially if you regular compare measurements from different machines. The "non medical" one might be several points off. It probably doesn't matter THAT much in the big picture, especially depending on how it is used (as a supplement).

      I agree there should be warnings on non-medically-approved devices for those treating dangerous medical conditions. As long as they are not trying to take them away.

      Looks like

  • by sabbede ( 2678435 ) on Thursday February 22, 2024 @11:13AM (#64259890)
    https://www.fda.gov/medical-de... [fda.gov]

    There, now you can see the actual FDA notice without pointlessly boosting ad revenue for schmucks.

  • by jerryjnormandin ( 1942378 ) on Thursday February 22, 2024 @01:03PM (#64260222)
    The Diabetes industry is so corrupt. I have type 2. I can better manage my glucose with intermittent fasting rather than taking Metformin and Glyburide every day. CGM sensors are expensive, even with insurance. I pay $79.00 for a 2 sensors that would last me 28 days. And that is with insurance. I tried a Chinese glucose watch and I calibrated it to a test with a test strip. The problem with the Chinese watch is it does not stay calibrated. If the glucose stays within a range of +-60mg/dl it's ok. If the you wake up at 200mg/dl, calibrate it for that.. If you spike to 260mg/dl or drop to 140mg/dl it will be off and you will need to recalibrate. I think if there was an OpenSource blood glucose monitoring watch I would be able to get it working. The FDA thinks end users are dumb. If I buy a glucose watch I would wear my CGM and test with blood strips for a month to be sure the watch is at least 10% within the reading from the CGM and blood strip test. The holy grail is an optical blood glucose sensor. Companies like Abbot are threatened since no one would need to purchase the overpriced Libre CGM . I may do intermittent fasting but I do monitor my glucose every day. 150mg/dl on no meds is way better than 100mg/dl on meds.
  • For good reason (Score:4, Informative)

    by hAckz0r ( 989977 ) on Thursday February 22, 2024 @01:58PM (#64260396)

    Some independent tests were done on many of the watches purchased off of Amazon and what they found was that these watches basically simulated the daily up/down trends of normal daily activity. They spiked during the times near breakfast/lunch/dinner and went down at preprogrammed times in between, modulus random variations. The readings actually had no relation to the true glucose levels in the blood but always gave readings that might make you think they were working correctly. These low end devices are a scam. Nothing more, and you are playing with your life if you try to depend on their accuracy.

    I was looking into this technology hoping that their might be something affordable that could give me a 24x7 reading but was quite dismayed with what I found about the fraud that exists on the current market. I'm still waiting for something that even gives near accurate readings without having to stick a probe into my skin. That's not happening any time soon it seems.

    • Well yeah. There is just outright FRAUD, like your example. And those should be sued or FTC'ed out of existence. And then there are the ones that do actually measure what they say, but not "medical use" accuracy. And those should be labeled appropriately.

      Uncovering a fraud should be pretty easy. If it is sold on Amazon or the like, I would expect that to show up pretty damn quickly. It is one reason I ALWAYS read the bad reviews on products.

  • A little off topic of measuring blood glucose levels: about 30 yrs ago had a professor who was working on control logic to automatically dispense diabetic medicine based on several real time measurements. Couldn't get it working due to so many nonlinear body systems (and each are person specific). I'm sure things have progressed since then, but it is not as straightforward as people might assume.
  • They would regulate the cost of insulin.

    The fact that the sky is the limit on what the price of insulin is should tell you all you need to know about how much the Government and their agencies truly care.

God doesn't play dice. -- Albert Einstein

Working...